A 1/2″ scotia moulding was stuck and applied round the top of the box carcase and a 1/4″ astragal was stuck and applied round its base.
There are two approaches to dovetailing the legs to the pillar. The first leaves the cylindrical base of the pillar in tact and the legs are dovetailed and coped to fit the radius of the pillar.
The second method is to cut three flat facets into the base of the pillar, each the same width as the legs. The dovetail sockets are then cut into these facets.
Either technique is appropriate for this period, but of the few pillar and claw leg joints I have had to repair, I’m fairly certain coped examples would out number the faceted ones which may or may not be significant; I can’t say. However, there are several pillar and claw tables in my home which I made with coped leg joints and I don’t ever want that to be interpreted by anyone as my preferred method or the only technique I employed.
Incidentally, iron braces (or triangles as they were then known) are not necessarily evidence of a repair; London cabinetmakers, at least, offered them as an option on new tables:
|A triangle on a pillar-and-claw table||0||0||2|
|Letting in a triangle plate, the sides not exceeding four inches long||0||0||4|
A reader asked me to provide a break down of the procedure of attaching the legs to the pillar, so here it is: I laid out the three centre lines (120° apart) for the legs’ sliding dovetail sockets and off-set the lines by half the width of the legs. I scribed a line from the intersection of one of each pair of these lines where it meets the circumference of the pillar across to their opposite line and pared off the waste, thus creating the facets.
The depths of the dovetails were marked off the faces of the facets with a marking gauge and then using a sliding bevel, I set out the dovetails onto the end of the pillar. The dovetail sockets were first sawn down as far as possible, and then chisels and mallet took over to complete the joints.
The same dimensions were laid out on the legs and the dovetails were then sawn and trimmed to shape.
I dry-assembled the pillar and legs and scribed the radius of the stop collar onto the tops of the legs. The legs were then withdrawn and trimmed.
The colour difference between pillar and legs in the image below is of no consequence; both the pillar halves and legs were cut from the same plank of walnut. The apparent dissimilarity is due to the pillar being burnished on the lathe with a handful of shavings and it therefore reflects more of its true colour. The legs have been sanded to 120-grit, so they absorb a greater amount of light. When wetted however, the colour is fairly uniform across pillar and legs.
Elm won the day for the block. I cut twin mortises through the block and tapered its ends so it will be less obtrusive when the table is complete.
I fettled the column to be a nice fit inside the pillar. It drops slowly on a cushion of air into the pillar and is impossible to withdraw swiftly due to the vacuum created, but makes a pleasing ‘pop’ when it is gradually pulled out. I will probably plane a little more off each of the column’s canted corners to allow the air an easier passage, otherwise raising the table will be a right royal pain!
A locating hole (into which the locking screw engages) was drilled in the top of the column so the table can be picked up by the carcase without it coming away from the pillar. Two subsequent holes were drilled at 2″ intervals down the column. I made two saw cuts in each of the tenons before gluing the column into the block. A wedge was hammered into each cut, splaying the edges of the tenons and locking the column securely into the block.
I eventually settled on apple for the locking screw. The wing is fairly typical of the decoration on wooden and brass screws used on furniture and architectural joinery throughout the Georgian period.
 The London Society of Cabinet-makers, The Cabinet-makers’ London Book of Prices, 1803, p. 266.